Creating an Incident Investigation Framework for a Complex Socio-Technical System: Application of Multi-label Text Classification and Bayesian Network Structure Learning

Reliability Engineering & System Safety, Journal Year: 2025, Volume and Issue: unknown, P. 110971 - 110971

Published: Feb. 1, 2025

Language: Английский

---

A New Era in Human Factors Engineering: A Survey of the Applications and Prospects of Large Multimodal Models

International Journal of Human-Computer Interaction, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 14

Published: Jan. 19, 2025

Language: Английский

---

Accident investigation via LLMs reasoning: HFACS-guided Chain-of-Thoughts enhance general aviation safety

Expert Systems with Applications, Journal Year: 2025, Volume and Issue: 269, P. 126422 - 126422

Published: Jan. 9, 2025

Language: Английский

---

Statistical Analysis of the Characteristics and Laws in Larger and Above Gas Explosion Accidents in Chinese Coal Mines from 2010 to 2020

Fire, Journal Year: 2025, Volume and Issue: 8(3), P. 87 - 87

Published: Feb. 21, 2025

Gas explosions are the most serious type of accident in coal mines China. This study analyzed 125 gas explosion accidents that occurred between 2010 and 2020. The results showed number deaths 2010–2020 was stable decreasing. larger is largest, but death toll from major much greater. Coal faces, headings, roadways main locations where initiated. which occur faces headings mainly “township” enterprises private mines, all engage illegal operations. cause accumulations ventilation system failure; these failures can be reduced with improved ventilations management. related Sichuan, Guizhou, Heilongjiang provinces very high. annual change frequency accidents, quarterly distribution time during a mining shift when closely to national policies regulations, company production goals, mental status miners, respectively.

Language: Английский

---

An Embodied Intelligence System for Coal Mine Safety Assessment Based on Multi-Level Large Language Models

Sensors, Journal Year: 2025, Volume and Issue: 25(2), P. 488 - 488

Published: Jan. 16, 2025

Artificial intelligence (AI), particularly through advanced large language model (LLM) technologies, is reshaping coal mine safety assessment methods with its powerful cognitive capabilities. Given the dynamic, multi-source, and heterogeneous characteristics of data in typical mining scenarios, traditional manual are limited their information processing capacity cost-effectiveness. This study addresses these challenges by proposing an embodied intelligent system for based on multi-level models (LLMs) multi-source sensor data. The employs a multi-layer architecture implemented multiple LLMs, enabling not only rapid effective but also enhanced environmental perception physical interactions. By leveraging tool invocation reasoning capabilities LLM conjunction knowledge base, achieves logical inference, anomalous detection, potential risk prediction. Furthermore, memory functionality ensures learning utilization historical experiences, providing solid foundation continuous processes. established comprehensive experimental framework integrating numerical simulation, scenario real-world testing to evaluate intelligence. Experimental results demonstrate that effectively processes exhibits rapid, efficient during interactions, offering innovative solution safety.

Language: Английский

---

Intelligent Gas Risk Assessment and Report Generation for Coal Mines: An Innovative Framework Based on GLM Fine-Tuning

Electronics, Journal Year: 2025, Volume and Issue: 14(2), P. 379 - 379

Published: Jan. 19, 2025

Traditional coal mine gas risk assessment relies on manual operations, leading to inefficiencies, incomplete information integration, and insufficient evaluation accuracy, ultimately affecting safety oversight. This paper proposes an intelligent report generation framework (IGRARG) based fine-tuning a Generative Language Model (GLM) address these challenges. The integrates multi-source sensor data with the reasoning capabilities of large language models (LLMs). It constructs dataset for scenarios, fine-tuned GLM. Incorporating industry regulations domain-specific knowledge base enhanced Retrieval-Augmented Generation (RAG) mechanism, automates alarm judgment, suggestion generation, creation via hierarchical graph structure. Real-time human feedback further refines decision making. Experimental results show accuracy 85–93%, over 300 field tests achieving 94.46% judgment reducing weekly from 90 min 2–3 min. significantly enhances intelligence efficiency assessment, providing robust support management.

Language: Английский

---

Theory and practice of solution strategies for unsafe acts based on accident causation models: A systematic review

Journal of Loss Prevention in the Process Industries, Journal Year: 2025, Volume and Issue: unknown, P. 105605 - 105605

Published: Feb. 1, 2025

Language: Английский

---

A comparative analysis for automated information extraction from OSHA Lockout/Tagout accident narratives with Large Language Model

Procedia Computer Science, Journal Year: 2025, Volume and Issue: 253, P. 1362 - 1372

Published: Jan. 1, 2025

Language: Английский

---

Prevention and control strategy of coal mine water inrush accident based on case-driven and Bow-Tie-Bayesian model

Energy, Journal Year: 2025, Volume and Issue: unknown, P. 135312 - 135312

Published: Feb. 1, 2025

Language: Английский

---

Functional evidential reasoning model (FERM) – A new systematic approach for exploring hazardous chemical operational accidents under uncertainty

Chinese Journal of Chemical Engineering, Journal Year: 2025, Volume and Issue: unknown

Published: March 1, 2025

Language: Английский

---